Your AI receptionist, set up for you in 48–72 hours (done-for-you)

Setting up an AI receptionist sounds like a brochure promise — "live in 48–72 hours" — until you see who does the work: not you. The scary part of any new tool — connecting the phone, plugging in WhatsApp, setting up the calendar, teaching the AI to sound like your business — we do that. You tell us how you work, hand us a few things, and approve the result before it talks to anyone real. That's what "done-for-you" means: two or three days after you start, you've already got the voice and chat agent and the CRM running, without your team touching anything technical.

We mean that literally, not as a slogan. Let's open the box and walk through exactly what happens between the first call and the moment the AI answers on its own.

What "done-for-you" really means

There are two ways to sell customer-service software. One is to hand you an empty dashboard, a pile of tutorials, and a good-luck wave: "set it up however you like." The other is to build it for you and hand it over working. Totem is the second, and the difference isn't convenience — it's whether the project ever launches or stalls half-done in the "I'll set it up next week" tab.

That means the technical part is ours. We connect your telephony so the AI answers without you changing your number; we link WhatsApp, Instagram, Messenger, SMS, email, and web chat into a single inbox; we sync your Google or Microsoft Calendar; and we build your Kanban-style lead board with the automation that moves each contact based on how the conversation went. You don't open a flow editor or wrestle with integrations. You tell us how you work; we turn that into an agent that works the way you do.

And one piece a lot of people assume doesn't exist: you approve. The agent doesn't talk to a real customer until you've read how it responds, heard how it sounds, and signed off. If you want it to say "practice" instead of "clinic," we change that before it goes live. The tone is yours; we just build it.

AI receptionist setup, step by step (48–72 hours)

The 48–72-hour range isn't magic: the setup is an implementation process with four short steps. Here's what happens, in order, from the moment you decide to try it to the moment the AI answers on its own.

1. 1

   Discovery: the kickoff call

2. 2

   Configuring the agent and the board

3. 3

   Testing with you

4. 4

   Live in 48–72 hours

The step that matters most is the third: testing. We don't ask you to trust blindly. You dial the number, call, and hear the AI respond; you text over WhatsApp and watch how it replies and how it logs the lead on the board. That rehearsal is where we fine-tune the grounding in your knowledge — so the AI answers with your information and doesn't make anything up — and where you decide which cases it handles on its own and which ones go to a person. If you want to dig into that part, we cover it in how to train your AI receptionist.

Once everything's approved, we turn on the channels and the AI starts answering: it picks up calls in ~1 second and replies over WhatsApp, Instagram, Messenger, SMS, and web chat right away, 24/7. The human part doesn't disappear — quite the opposite: any conversation hands off to a person in one click, with all the context of what the customer has already said. The AI answers first; your team steps in where they add value.

What we need from you for setup (and what we don't)

The honest question before you start is "how much work is going to land on me during setup?" Short answer: not much, and nothing technical. Here's what we do ask of you and what we don't.

For that discovery, we just need material you already have: your contact channels, what your calendar looks like, the questions customers ask you most, and, if you have them handy, a couple of documents with your services or prices. You don't need to prepare anything elaborate or write scripts — we handle that from what you tell us. You don't switch phone numbers or phone systems either — the AI answers on what you already have, as we explain in how to put an AI on your calls without changing your number.

And because Totem is built to the EU's AI Act and GDPR — a premium, world's-strictest baseline — we set the compliance pieces up by default, not as an extra you have to assemble yourself. If you take calls from or handle data of EU residents, those rules apply to you directly; either way, you get that standard out of the box. The agent is designed to disclose that it's an AI when appropriate: it's the AI Act's transparency duty (Regulation (EU) 2024/1689, art. 50.1). As a UX best practice — not because GDPR requires it for a booking — we always leave a way for the customer to escalate to a person; GDPR's so-called "right to human intervention" only arises for solely automated decisions with legal or significant effects (art. 22 GDPR), and a booking or reception agent normally doesn't make that kind of decision. And on where the data lives: GDPR doesn't require hosting it in the EU, only applying safeguards when it leaves the EEA (Regulation (EU) 2016/679, Ch. V, art. 44) — which is why we work with providers and adequate transfer safeguards. In the US, the domestic expectations come from the FTC and, for calls and texts, the TCPA. It's not paperwork that lands on your team; it's part of the setup. This is information, not legal advice: when in doubt, consult your DPO or a lawyer. If you want the legal detail, we break it down in whether an AI chatbot or voice agent is legal for customer service.

No long-term contract: why we bet on the outcome

Here's the part that ties everything together. Done-for-you onboarding is included, with no setup fee, and there's no long-term contract. You pay month to month, and if at any point it's not a fit, you leave whenever you want. That's not generosity — it's consistency. If we did the setup and then locked you into a twelve-month contract, we'd be covering ourselves in case the results don't show up. We don't cover ourselves. The outcome is our problem.

That's the logic behind investing 48–72 hours of our team's time to get you live: if the agent doesn't bring you more bookings, doesn't stop missed calls, and doesn't deliver a board that genuinely moves your leads, nothing keeps you here. We lay it all out in no-contract AI receptionist: cancel anytime. For context, according to our clients Totem can cut response time by around 60%, generate about 40% more demos, and reduce no-shows by roughly 35% — these aren't guarantees, they're the ranges we see when the setup is done right, which is exactly why we put those hours in.

The takeaway is simple: the decision you're making isn't "I'm going to build myself an AI customer-service system." It's "I'm going to let Totem build it and hand it to me working in two or three days." Your team doesn't turn into integration engineers; they keep serving their customers, now with an AI that answers first and a board that never lets a lead go cold.

Official sources

• AI Act (Regulation (EU) 2024/1689), art. 50 (transparency) and art. 113 (application timeline) — EUR-Lex
• GDPR (Regulation (EU) 2016/679), art. 22 (automated decisions) and Ch. V, art. 44 (international transfers) — EUR-Lex